Patent Application
20240110880
Charged particle microscope systems have been developed to allow scientists to investigate and gather information on how microscopic systems work. In pursuit of such knowledge, scientists push the limits of what current charged particle microscope systems are able to investigate. One fundamental limitation that charged particle microscopy faces is the difficulty of imaging or otherwise investigating samples composed of materials that are highly reactive or otherwise susceptible to damage from charged particle or electron beams.
This is especially a problem when scientists attempt to measure the chemical context (e.g., oxidation state, ionization, valence state, etc.) of elements using analytical techniques such as EELS, as such techniques require an excessively long dwell time to obtain precise chemical information. Using present techniques, current charged particle systems struggle to obtain high resolution chemical contexts for samples before the cumulative damage caused by irradiation from the system's charged particle or electron beam destroys the material being investigated. This problem is exacerbated for samples composed highly reactive materials, such as batteries, as the beam dose that such reactive materials can receive before being damaged is much lower. Accordingly, there is desired to have new systems and methods for investigating samples that are able to acquire the chemical context of sample materials with reduced beam dosage requirements.
Methods for using dynamic data-driven detector tuning to investigate a sample with a charged particle microscopy system according to the present disclosure include acquiring sample data for a region of interest on the sample, and then determining one or more materials present in the region of interest. Once the materials are identified, a differentiation detector window is identified for the one or more materials, and the detector settings of a detector are adjusted such that the detector obtains information within the differentiation detector window. Thus, as the sample is subsequently scanned, the detector obtains an optimal range of information that is allows for efficient differentiation among the one or more materials.
Systems for investigating a sample using dynamic data-driven detector tuning to according to the present disclosure may comprise a sample holder configured to hold a sample, a charged particle source configured to emit a beam of charged particles towards the sample, an optical column configured to cause the beam of charged particles to be incident on the sample, and one or more detectors configured to detect charged particles of the charged particle beam and/or emissions resultant from the charged particle beam being incident on the sample. According to the present disclosure the one or more detectors include an adjustable detector having adjustable detector settings, wherein the detector settings of the adjustable detector can be changed so that the adjustable detector obtains information within a desired differentiation detector window. The systems also include one or more processors, and a memory storing computer readable instructions that, when executed by the one or more processors, cause the corresponding system to perform one or more steps of methods according to the present disclosure.
The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identify the figure in which the reference number first appears. The same reference numbers in different figures indicates similar or identical items.
Like reference numerals refer to corresponding parts throughout the several views of the drawings. Generally, in the figures, elements that are likely to be included in a given example are illustrated in solid lines, while elements that are optional to a given example are illustrated in broken lines. However, elements that are illustrated in solid lines are not essential to all examples of the present disclosure, and an element shown in solid lines may be omitted from a particular example without departing from the scope of the present disclosure.
Methods and systems for data driven detector tuning for charged particle systems, are included herein. More specifically, the methods and systems disclosed herein include and/or are configured to determine a detector window that allows for efficient (i) differentiation of materials, and/or (ii) chemical context of such materials in sample, and then adjust the settings of a detector in a charged particle system to allow for data to be collected within the determined window. In this way, the methods and system of the present invention are able to use information and/or images of a sample region to determine materials and/or material characteristics that are potentially present within the sample region, and then dynamically determine an optimal differentiation detector window such that the detector can obtain sufficient data to allow for subtle material/material characteristics differentiation with a very low dose beam. According to the present disclosure, the differentiation detector window corresponds to detector settings that result in the corresponding detector capturing sample data of a desired characteristic range (e.g., intensity range, energy resolution, energy range, etc.) that includes a differentiating feature that allows a material or a characteristic of the material present in the sample to be identified out of one or more potential materials or potential characteristics of the materials. For example, a potential material may correspond to any of a type of compound, a chemical, an element, an ionization state, an oxidation state, a plasmon, a plasmon peak, a phonon, a valence state, etc. Because prior systems use less efficient set-ups of detectors, these prior systems required high beam dosages before detectors could obtain sufficient data to make such subtle differentiations, some materials and material characteristics have been difficult or impossible to determine due to the degradation of the sample due to beam irradiation. However, since the differentiation detector window optimally causes the detector system specifically captures a characteristic range of sample data that includes a differentiating feature, the systems and methods of the present disclosure are able identify sample materials with lower beam dosages and/or dwell times.
That is, the present system makes an initial determination on the potential materials/material characteristics of the sample, then uses this determination to identify the differentiations that are necessary to identify materials/material characteristics, and then dynamically determines and sets the detector window to most efficiently capture detector data indicative of the identified differentiations, the methods and systems of the present invention are able to quickly differentiate and identify subtle material/material characteristics. Additionally, the methods and system of the present disclosure may determine a plurality of detector windows that each relate to the most efficient detector settings to obtain data for the differentiation between corresponding material/material characteristics, such that multiple material/material characteristics can be quickly determined within a sample region without damaging the sample.
In various embodiments, the charged particle beam may be scanned one or more times across an entire region with each determined detector window to obtain the desired detector data for the corresponding differentiation. Alternatively, the charged particle beam may only scan particular regions of a sample region that are expected to have a potential material/material characteristic for each corresponding detector window (e.g., changing the detector window to the optimal settings at each pixel or point of the sample, scanning a plurality of pixels/sample points with a same setting before changing the detector window and scanning a new set of pixels/sample points, etc.) to further reduce the time to results and sample damage due to irradiation. Not only does this reduce time to results for operators, but it also allows for new materials/material characteristics to be identified while reducing sample damage. According to the present disclosure, the methods and systems may
The example charged particle microscope system(s) 104 includes a charged particle source 108 (e.g., a thermal electron source, Schottky-emission source, field emission source, etc.) that emits an electron beam 110 along an emission axis 112 and towards an accelerator lens 114. The emission axis 112 is a central axis that runs along the length of the example charged particle microscope system(s) 104 from the charged particle source 108 and through the sample 102. The accelerator lens 114 that accelerates/decelerates, focuses, and/or directs the electron beam 110 towards a focusing column 116. The focusing column 116 focuses the electron beam 110 so that it is incident on at least a portion of the sample 102. In some embodiments, the focusing column 116 may include one or more of an aperture, scan coils, and upper condenser lens. The focusing column focuses electrons from electron source into a small spot on the sample. Different locations of the sample 102 may be scanned by adjusting the electron beam direction via the scan coils. Additionally, the focusing column 116 may correct and/or tune aberrations (e.g., geometric aberrations, chromatic aberrations) of the electron beam 110.
The STEM system 106 is further illustrated as having a projector lenses/projector system 120 that receive the portions of the charged particle beam 110 that transmit through the sample 102. Electrons 128 scattered by the sample 102 may be recorded by a STEM detectors 126, and/or may enter the EELS spectrometer system 107. The spectrometer system 107 comprises a dispersive element 130 which fans out the electrons to a spectrum, and a system of lenses 132 which magnifies the spectrum to a magnified spectrum at detector 134.
The computing device(s) 124 are configured to control operation of the example charged particle microscope system(s) 104, generate images of sample 102 and/or otherwise determine or interpret data from the detector systems 122, 126, and 134. According to the present invention, the computing device(s) 124 are configured to cause the charged particle microscope system(s) 104 to irradiate one or more locations on a surface region of the sample 102 with charged particle beam 110 (e.g., an electron beam), obtain detector data from a detector system 122, 126, and 134 (e.g., a dark field, EELS, EDS, EDX, XEDS or other type of imaging detector system), and then generate sample information (e.g., spectrum image, energy loss spectrum, a diffraction pattern, an initial image of the surface region, etc.) based on the detector data. The computing device(s) 124 are further configured to identify sample characteristics for regions on the surface within the initial sample information.
According to the present disclosure, the computing device(s) 124 are configured to initially acquire sample data for a region of interest on a surface of the sample 102. The acquired sample data may correspond to an image of the region, an EDX scan of the region, a STEM scan of the region, an EELS scan of the region, a low resolution EELS scan of the region, a low resolution EDX scan of the region, or another type of data from which the computing device 124 can determine potential materials present in the region. A scan may be configured to consist of regular path in which the beam is scanned line by line over the sample, or in a less regular path such as a Hilbert scan or a sparse scan. A scan path may be configured to probe every point on the sample or to probe only a limited sub-region or limited number of (possibly arbitrarily distributed) points on the sample. Furthermore, a scan may be configured as a ‘single scan’ in which every scanned point on the sample (or, equivalently, every pixel in the resulting scan image) is visited once, or as a ‘multiple scan’ in which every point on the sample is visited multiple times. In some embodiments, the computing device(s) 124 may cause the charged particle microscope system(s) 104 to acquire the initial sample data for the region of interest on a surface of the sample 102. Such an initial acquisition of the sample data can be done rapidly and with a low dose of charged particle beam irradiation, as the initial sample data is used by the computing device(s) 124 to identify sample materials that are potentially present in the region of interest. Alternatively, or in addition, the computing device(s) 124 may obtain some of all of the initial sample data from a network, a hardware connection, an accessible memory, and/or user input.
The computing device(s) 124 are then configured to determine one or more materials that are likely to be present in the region of the sample 102 based on acquired the initial sample data. Due to the low dose required to identify materials in the sample using the systems and methods of the present disclosure, the types of materials that can be differentiated and/or identified include material types that have previously been difficult or impossible to be differentiated and/or identified. For example, potential materials that can be identified as being potentially present in the sample (and subsequently being differentiated/identified) include but is not limited to types of compounds, types of chemicals, elements, oxidations states of materials, plasmons, plasmon peaks, valence states of atoms, etc.
In various embodiments, the materials that are potentially present in the region of the sample 102 is determined based on the sample data that has been initially acquired. In some embodiments, the computing device(s) 124 may use the sample data to identify materials that are potentially present in the sample by assigning a score to the likelihood that the initial sample data indicates the presence of the material, and then select a set of one or more potential materials who have the highest score (i.e., the highest three scoring materials), that exceed a threshold value, or a combination thereof. For example, where the initial sample data includes a low-resolution EELS scan of the region, the computing device 124 may identify one or more materials that have a spectral fingerprint that is similar to the spectrum shown information shown in the initial sample data. A spectral fingerprint corresponds to the spectral information that is obtained from an investigation of the corresponding material. In such examples, the computing device 124 may assign a similarity score between a plurality of spectral fingerprints and the spectrum information of the initial sample information, and then determine one or more materials that are potentially present in the region of the sample based on the scores.
Alternatively, or in addition, the computing device(s) 124 may receive a user input and/or access a data structure that identifies potential materials that could be present in the type of sample, groupings of materials that are commonly confused with each other, and/or materials of interest, and then select the one or more potential materials based on the information in the data structure. For example, based on receiving a user input of a type of sample being investigated and a material of interest, the computing device(s) 124 may access a data structure that identifies a grouping of materials that are likely to be found in the sample type and/or are likely to be misidentified as the material of interest. The computing device(s) 124 may then use this information in combination with the initial sample data to determine the one or more materials that are likely to be present in the region of the sample 102.
In various embodiments, the computing device(s) 124 may further use the initial sample information to identify additional regions of the sample where the one or more materials (and/or characteristics) are likely to be present. For example, based on the initial sample data for the additional regions of the sample having similar information (e.g., similar spectrum, image of the surface of the sample 102 showing similar features/structures, etc.) the computing device(s) 102 may determine that the sample materials are also likely to be present in the additional regions. Alternatively, or in addition, the computing device(s) 124 may independently repeat the process of identifying potential materials based on initial sample data for multiple regions of the sample 102. In this way, the computing device(s) 124 can identify a plurality of sets of one or more potential materials, with each set corresponding to one or more regions of the sample 102.
Once one or more materials (and/or characteristics) that are potentially present in the region of interest of the sample 102 are determined, the computing device(s) 124 then determines a differentiation detector window for the one or more materials that is configured to allow for the detector 122, 126, or 134 to capture sample data of a desired characteristic range (e.g., intensity range, energy resolution, energy range, etc.) that includes a differentiating feature that allows a material or a characteristic of the material present in the sample to be identified out of the one or more potential materials or potential characteristics of the materials. For example, a differentiation windows may correspond to an energy range that is quantized by the detector system 122, 126, or 134, a quantized intensity range (e.g., only capturing detector input with a saturation between 500 and 1200 counts), or other detector settings that result in the capture an energy resolution of a desired range.
As is understood by people skilled in the art, most types of charged particle systems 104, 106, 107 have settings that may be adjusted to enable the system to optimally identify or record the materials (or material characteristics) present in the sample, how materials/material characteristics are distributed across the sample, and/or specific details of one or more particular materials (or material characteristics). For example, the EELS spectrometer 107 includes lenses 132 that can be adjusted to image a large range of an EELS spectrum with moderate energy resolution on detector 134 when in a first alignment, and image a small range of an EELS spectrum with high energy resolution image on detector 134 (e.g., the finite number of pixels in detector 134 limits the simultaneously attainable energy range and energy resolution) when the lenses 132 are in a second alignment. A person skilled in the art would understand that first alignment is optimal for recording EELS spectrum overviews which reveal in one exposure which elements are present in the specimen, while the second alignment is optimal for recording the fine details inside a specific EELS (e.g., a peak that reveals chemical information such as oxidation state). While alignment of lenses 132 is one way in which detector settings of the EELS spectrometer 107 can be adjusted, other example ways of adjusting the detector settings includes, but are not limited to, changing the exposure time per spectrum (e.g., relatively short exposure times are sufficient for establishing just the presence of an EELS peak, thus the presence of a specific element, whereas longer exposure time is needed to collect sufficient signal to properly discern the fine details in an EELS peak), the sub-range of the EELS spectrum which is recorded when the spectrometer is set to record at a high energy resolution, etc.
As another example, in diffraction imaging, the range of the diffraction pattern captured by the detector (typically a 2D image sensor) may be adjusted by changing the lenses of the projection system 120. For example, in the so-called 4D-STEM method, a probe of a few nanometer scans over the sample 102, and a 2D diffraction pattern is recorded at each position (thus generating a 4D data set). When the lenses of the projection system 120 are set to relatively low magnification of the diffraction pattern, the resulting 4D-STEM data-set may be optimal for identifying where the various crystal structures possibly present in the sample are located. Alternatively, when the lenses of the magnifying system are set to relatively high magnification of the diffraction pattern, the resulting 4D-STEM data-set may be optimal for quantifying distortions in the lattice structure (due to, for example, stress or dopants). Persons having skill in the art know that other settings of the 4D-STEM experiment set-up that can be adjusted are exposure time per diffraction image, or which sub-area of the diffraction pattern is recorded when the magnifying system is set to record at high magnification of the diffraction pattern.
In the above examples, the settings of the detector may be adjusted such that the detector selects a specific window of a spectrum or image, such that, in this specific window, a sufficient signal and/or resolution is available to enable differentiation between the one or more particular details in material and/or material characteristics desired by an operator. We will call such setting which enables such differentiation the ‘differentiation detector window’ (DDW). As is clear from above examples, and as those skilled in the art know, the process of recording a data set with the corresponding DDW settings is much more time consuming than merely recording a data set with settings that are used to identify which materials are generally present in the sample and/or how they are distributed across the sample. It is therefore advantageous and desirable to limit acquisitions using the DDW settings only to regions where the particular materials/material characteristics to which the DDW is optimal are located (or which are potentially located) on the sample.
It is also noted that the settings which can be adjusted to define the DDW do not necessarily only comprise parameters as magnification and sub-range of spectrum or image, but can also comprise settings other such as detector amplification, detector brightness, and detector bias. Typically, all detector settings that cannot be switched near-instantaneously (i.e., within a time significantly shorter than the time required for one detector read-out) can be part of the set of settings defining the DDW.
In some embodiments, the computing device(s) 124 may use preset differentiation detection windows for differentiating between particular materials. Alternatively, the computing device(s) 124 may compare spectral fingerprints associated with each of the potential materials and identify spectral ranges and/or detector windows that capture at least one difference that allows the materials to be differentiated (e.g., by accessing a library of expected spectra for different materials of interest). For example, where two spectral fingerprints are nearly identical except for particular portion spectral range that includes a peak in the spectral fingerprint for a first potential material a subtle double peak in the spectral fingerprint for a second potential material, the computing device(s) 124 can select a differentiation detector window that corresponds to the particular portion of the spectral range.
In some embodiments, the computing device(s) 124 are configured to adjust the detector settings of a detector 122 to cause the detector to obtain information within the differentiation detector window. In this way, when the detector 122 is configured to acquire sample information within this differentiation detector window, a user and/or algorithm would be able to quickly determine what material is actually present in the region of the sample by determining whether a single peak or a double peak occurs within the sample data. For example, where the detector 122 is an EDX detector, the detector settings may cause the detector to capture sample data that corresponds to X-rays having energies within a range that includes a differentiating feature between one or more potential materials. In another example, where the detector 122 is a pixelated detector, the detector settings may cause the detector to capture sample data with a specific camera length that allows for the visualization of a part of the diffraction space that includes a differentiating feature between one or more potential materials. Because the detector 122 is only capturing sample data within the differentiation detector window, the amount of data needed to identify the material present in the sample is greatly reduced, allowing a lower power charged particle beam to be used to irradiate the sample, a shorter amount of time spent irradiating the sample, or both. In this way, the systems and methods of the present disclosure are able to efficiently and quickly identify the materials present in a sample with much improved time to results and greatly reduced sample damage due to beam dose.
The computing device 124 may then cause the charged particle microscope system(s) 104 to initiate an investigation of the corresponding region of the sample 102. In various embodiments, the computing device 124 may cause the charged particle microscope system(s) 104 to scan a single region of the sample 102, may scan a plurality of regions of the sample 102 that have been determined to have the same materials potentially present, or may scan an entire surface of interest with the detector 122 being configured to obtain sample data within the differentiation detector window.
Additionally, in embodiments where the computing devices 124 have determined multiple differentiation detector windows, the computing devices 124 may cause the settings of the detector 122 to be adjusted to reflect an additional detector window, and then cause the charged particle microscope system(s) 104. In some cases, the new differentiation detector window may have been determined to capture a difference in the spectral fingerprint for two new potential materials in the same and/or a different region of the sample 102. Alternatively, the new differentiation detector window may have been determined to capture a difference in the one or more potential materials associated with a previous differentiation detector window which still remain to be differentiated by a previous scan (e.g., the previous scan had insufficient data to identify an individual material, a subset of potential material still needs to be differentiated, etc.).
The computing devices 124 may cause the charged particle microscope system(s) 104 to scan the sample with the detector 122 being configured to obtain sample data within multiple different differentiation detector windows. For example, the charged particle microscope system(s) 104 may first scan a first plurality of regions of the sample 102 associated with first potential materials used to determine a first differentiation detector window, and then scan a plurality of second regions of the sample 102 associated with second potential materials used to determine a second differentiation detector window. Alternatively, the charged particle microscope system(s) 104 may cause the charged particle beam 110 to scan along the surface of the sample 102 along a known path, and the computing devices 124 may cause the detector settings of the detector 122 to dynamically change to the determined differentiation detector window for the corresponding region of the sample 102 that is currently being/about to be scanned.
Then, because the sample data obtained by the detector 122 is within the differentiation detector window that was selected to capture only sample data that differentiations between potential materials in the sample, the computing device 124 is able to then determine the material that is actually present in a region of the sample 102 based on the sample data obtained from low dose irradiation of the region. For example, according to the present invention, where the differentiation detector window was determined based on potential materials corresponding to a plurality of valance states for a particular atom, the computing device 124 may identify the valence states of individual atoms as they are scanned with a low dose beam.
Those skilled in the art will appreciate that the computing devices 124 depicted in
It is also noted that the computing device(s) 124 may be a component of the example charged particle microscope system(s) 124, may be a separate device from the example charged particle microscope system(s) 104 which is in communication with the example charged particle microscope system(s) 104 via a network communication interface, or a combination thereof. For example, an example charged particle microscope system(s) 104 may include a first computing device 124 that is a component portion of the example charged particle microscope system(s) 104, and which acts as a controller that drives the operation of the example charged particle microscope system(s) 104 (e.g., adjust the scanning location on the sample 102 by operating the scan coils, causes translations of the sample 102, etc.). In such an embodiment the example charged particle microscope system(s) 104 may also include a second computing device 124 that is desktop computer separate from the example charged particle microscope system(s) 104, and which is executable to process data received from one or more detector system(s) 122 to generate images of the sample 102, determine scan strategies for the sample 102, and/or perform other types of analysis. The computing devices 124 may further be configured to receive user selections via a keyboard, mouse, touchpad, touchscreen, etc.
In various examples, such initial sample information may include an image, an EDX scan, a STEM scan, an EELS scan, a low resolution EELS scan, a low resolution EDX, or another type of data from which a set of potential materials located within the sample can be identified. For example, based on a quick low-resolution EELS scan the computing devices 102 may identify portions of the sample from which similar spectral information was obtained, access a set of spectral fingerprints for different materials, and then select two or more materials having similar spectral fingerprints as potentially being located within the corresponding regions of the sample.
Alternatively, or in addition, the initial sample information may include a user input or sample identifier that provides potential materials, sample types, investigation types, expected sample compositions, etc. For example, based on reception of a sample identifier the computing device(s) 124 may determine a set of materials that are expected to be in the sample 102, then may access an image and/or spectral information from an initial scan of the sample 102 to identify portions of the sample that have characteristics which are similar to certain ones of the expected materials (e.g., image show formations having traits, image shows atom size within range for particular materials, spectral information is similar to the spectral fingerprint, etc.).
Image 144 shows a plurality of spectra fingerprints 152 that have been determined to be similar to the spectral information obtained in the initial sample data. Specifically, spectral fingerprints 152a and 152b correspond to spectral fingerprints for a first potential material and a second potential material that are similar to the spectral information obtained from the locations where the large central structure and the upper rightmost structure as depicted in image 142, and spectral fingerprints 152c and 152d correspond to spectral fingerprints for a third potential material and a fourth potential material that are similar to the spectral information obtained from the locations where deposits where depicted within the sample 102 in image 142. The spectral fingerprints may be identified based on a similarity to an initial sample data (e.g., a similarity score), may be based on a cluster of commonly confused materials, may be user selected, etc.
Image 146 shows an updated version of image 144 in which differentiation detector windows 154 have been determined. Specifically, image 146 shows a first differentiation detector window 154a that captures a differentiating feature between the first spectral fingerprint 152a and the second spectral fingerprint 152b, and a second differentiation detector window 154b that captures a differentiating feature between the third spectral fingerprint 152c and the fourth spectral fingerprint 152d. Image 146 shows that differentiation detector window 154a was selected to include a portion of the spectral fingerprints of the potential materials that was both easily differentiable from the third and fourth spectral fingerprints 152c and 152d (i.e., which have no contribution within the window), while also capturing spectral information that efficiently allows the first and second fingerprints 152a and 152b to be differentiated (i.e., the first spectral fingerprint 152a has a single peak within the window while the second spectral fingerprint 152b has a double peak within the window). In various embodiments the differentiation detector window may be determined based on a comparison between the spectral fingerprints of potential materials to identify the differentiating features, or may be based on a preset list of windows for particular sets of materials.
Image 148 illustrates an updated version of image 142 in which the regions of interest on the sample 102 have been scanned with the detector system 122 having been configured to the determined differentiation detector window 154 such that data was obtained that was able to identify the actual material present at the location from the predetermined set of potential materials. Because this scan is acquiring very specific sample data within a window to efficiently make differentiation decisions, the beam dose required to get sufficient sample data to make a material identification is greatly reduced. This allows the systems and methods of the present disclosure to differentiate and identify materials that were previously difficult or impossible to perform. For example, image 148 shows a first set of regions 156 of the sample that were identified as being a first type of element, a second set of regions 158 of the sample that were identified as being a second type of element.
In the example computing architecture 160, the computing device includes one or more processors 162 and memory 164 communicatively coupled to the one or more processors 162. The example computing architecture 160 can include a control module 166, a potential material determination module 168, a differentiation detector window module 170, a scan strategy determination module 172, and a composition determination module 174 stored in the memory 164.
As used herein, the term “module” is intended to represent example divisions of executable instructions for purposes of discussion, and is not intended to represent any type of requirement or required method, manner or organization. Accordingly, while various “modules” are described, their functionality and/or similar functionality could be arranged differently (e.g., combined into a fewer number of modules, broken into a larger number of modules, etc.). Further, while certain functions and modules are described herein as being implemented by software and/or firmware executable on a processor, in other instances, any or all of modules can be implemented in whole or in part by hardware (e.g., a specialized processing unit, etc.) to execute the described functions. As discussed above in various implementations, the modules described herein in association with the example computing architecture 160 can be executed across multiple computing devices 124.
The control module 166 can be executable by the processors 162 to cause a computing device 122 and/or example charged particle microscope system(s) 104 to take one or more actions. For example, the control module 174 may cause the example charged particle microscope system(s) 104 to scan a surface of the sample 102 by causing the charged particle beam 110. The computing device 112 may then be configured to generate an initial image of the surface of the sample 102 based on detector data from a detector system 122 obtained when the surface of the sample 102 is scanned. In an alternate example, the control module 174 may cause the configuration settings of the detector system 122 to be altered such that they obtain sample information within a differentiation detector window, and/or cause the example charged particle microscope system(s) 104 to scan a surface of the sample 102 according to a scanning strategy determined by the computing devices 124 based on the initial image.
The potential material determination module 168 can be executable by the processors 162 to obtain initial sample data for a region of interest on a surface of the sample 102. The acquired sample data may correspond to an image of the region, an EDX scan of the region, an integrated differential phase contrast (IDPC) scan of the region (a STEM method in which the STEM detector is segmented in at least 4 segments, and the difference in signals between opposite segments is recorded. This difference, when integrated, represents the charge distribution in the sample), an EELS scan of the region, a low resolution EELS scan of the region, a low resolution EDX scan of the region, or another type of data from which the potential material determination module 168 can determine potential materials present in the region. The potential material determination module 168 is further configured to determine one or more materials that are likely to be present in the region of the sample 102 based on the initial sample data. Example potential materials that can be identified as being potentially present in the sample (and subsequently being differentiated/identified) include but is not limited to types of compounds, types of chemicals, elements, ionization states of materials, plasmons, plasmon peaks, valence stats of atoms, etc.
In various embodiments, the materials that are potentially present in the region of the sample 102 is determined based on the sample data that has been initially acquired. In some embodiments, the potential material determination module 168 may use the sample data to identify materials that are potentially present in the sample by assigning a score to the likelihood that the initial sample data indicates the presence of the material, and then select a set of one or more potential materials who have the highest score (i.e., the highest three scoring materials), that exceed a threshold value, or a combination thereof. Alternatively, or in addition, the potential material determination module 168 may receive a user input and/or access a data structure that identifies potential materials that could be present in the type of sample, groupings of materials that are commonly confused with each other, and/or materials of interest, and then select the one or more potential materials based on the information in the data structure.
In various embodiments of the present disclosure, the potential material determination module 168 may identify the potential materials within a region interest using one or more of an identification algorithm, a machine learning module (e.g., an artificial neural network (ANN), convolutional neural network (CNN), Fully Convolution Neural Network (FCN) etc.) trained to identify instances of one or more structures of interest, user selections, and/or a combination thereof. For example, a neural network may be employed to segment the initial image according to whether the corresponding pixels contain a material of interest. Where the potential material determination module 168 segments the initial image into a plurality of different types of regions (i.e., regions likely to have a material of interest, regions having a features indicative of materials, etc.) a user may select individual types of regions that are to be investigated and/or select from a list of materials as potentially being present there. In an alternate example, the potential material determination module 168 may apply an algorithm to such a segmented image that is programmed to select regions to be investigated based on the characteristics/features of the regions (e.g., size, shape, location, proximity to other regions, etc.), and assign potential materials located in the regions based on the characteristics/features.
The differentiation detector window module 170 can be executable by the processors 162 to determine an optimal differentiation detector window for the one or more potential materials identified by the potential material determination module 168. In some embodiments, the differentiation detector window module 170 may use preset windows for differentiating between particular materials. Alternatively, the differentiation detector window module 170 may compare spectral fingerprints associated with each of the potential materials and identify spectral ranges and/or detector windows that capture at least one difference that allows the materials to be differentiated (e.g., by accessing a library of expected spectra for different materials of interest). For example, where four potential materials have been identified for a particular reason, the differentiation detector window module 170 may compare the spectral fingerprints of each potential material to find a common window for which each of the potential materials may be identified or verified as not present based on sample data obtained in from region. In some embodiments, the differentiation detector window module 170 may further be executable to adjust the detector settings of a detector 122 to cause the detector to obtain information within the differentiation detector window.
The scan path determination module 172 can be executable by the processors 162 to identify a scan strategy that is especially customized to the region of the sample 102 being investigated. Specifically, the scan path determination module 172 is executable to identify a beam path that the charged particle beam 110 is going to follow when scanning the sample 102. The beam path may correspond to a preprogrammed beam path that is constant for all samples or may be individual customized to the sample being investigated. For example, the scan path determination module 172 may determine a beam path that first irradiates the regions of the sample determined to have the same potential materials and/or are associated with the same differentiation detector window, and then—after adjusting the detector 122 to have a new differentiation detector window—irradiates a second set of regions of the sample which have been determined to have a new set of potential materials and/or have been associated with the new differentiation detector window. Alternatively, the scan path determination module 172 may assign differential detector windows to corresponding portions of the scan path, such that when the sample is investigated, the detector 122 settings are dynamically adjusted so that it has the appropriate differentiation detector window when the corresponding region of the sample is being scanned.
The composition determination module 174 can be executable by the processors 162 to determine compositional information about the materials present in particular regions of interest based on the sample information obtained by the detector system 122 when said region was scanned with the charged particle beam 110 while configurations of the detector system 122 are such that the detector window corresponds to the differentiation detector window determined for said region. For example, based on the sensor data obtained for a specific region, the composition determination module 174 may analyze the sensor data to see if identifying features for one of the potential materials is present. In this way, the composition determination module 174 is able to determine that a potential material is actually present, eliminate the potential that a material is present, determine if another scan is necessary (i.e., insufficient or erroneous sample data for the region), and/or if the process needs to be repeated (e.g., some but not all of the potential materials have been eliminated, and a new differentiated detector window needs to be determined).
As discussed above, the computing devices 124 include one or more processors 162 configured to execute instructions, applications, or programs stored in a memory(s) 164 accessible to the one or more processors. In some examples, the one or more processors 162 may include hardware processors that include, without limitation, a hardware central processing unit (CPU), a graphics processing unit (GPU), and so on. While in many instances the techniques are described herein as being performed by the one or more processors 162, in some instances the techniques may be implemented by one or more hardware logic components, such as a field programmable gate array (FPGA), a complex programmable logic device (CPLD), an application specific integrated circuit (ASIC), a system-on-chip (SoC), or a combination thereof.
The memories 164 accessible to the one or more processors 162 are examples of computer-readable media. Computer-readable media may include two types of computer-readable media, namely computer storage media and communication media. Computer storage media may include volatile and non-volatile, removable, and non-removable media implemented in any method or technology for storage of information, such as computer readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disk (DVD), or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that may be used to store the desired information and which may be accessed by a computing device. In general, computer storage media may include computer executable instructions that, when executed by one or more processing units, cause various functions and/or operations described herein to be performed. In contrast, communication media embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal, such as a carrier wave, or other transmission mechanism. As defined herein, computer storage media does not include communication media.
Those skilled in the art will also appreciate that items or portions thereof may be transferred between memory 164 and other storage devices for purposes of memory management and data integrity. Alternatively, in other implementations, some or all of the software components may execute in memory on another device and communicate with the computing devices 124. Some or all of the system components or data structures may also be stored (e.g., as instructions or structured data) on anon-transitory, computer accessible medium or a portable article to be read by an appropriate drive, various examples of which are described above. In some implementations, instructions stored on a computer-accessible medium separate from the computing devices 124 may be transmitted to the computing devices 124 via transmission media or signals such as electrical, electromagnetic, or digital signals, conveyed via a communication medium such as a wireless link. Various implementations may further include receiving, sending or storing instructions and/or data implemented in accordance with the foregoing description upon a computer-accessible medium.
At 202, an initial sample data for a sample to be investigated is acquired. The acquired sample data may correspond to an image of the region, an EDX scan of the region, an IDPC scan of the region, an EELS scan of the region, a low resolution EELS scan of the region, a low resolution EDX scan of the region, or another type of data from which the potential material determination module 168 can determine potential materials present in the region. In various embodiments, the initial sample data may be acquired by scanning the sample with a charged particle device (e.g., electron microscope), an imaging device, user input, accessed via an accessible memory, over network connection, or a combination thereof.
At 204, one or more potential materials are determined to potentially be present in a region of interest on the sample based at least in part on the initial sample data. Example potential materials that can be identified as being potentially present in the sample (and subsequently being differentiated/identified) include but is not limited to types of compounds, types of chemicals, elements, ionization states of materials, plasmons, plasmon peaks, valence stats of atoms, etc.
In various embodiments, the potential materials may be determined by an algorithm that is programed and/or trained to identify potential materials in a region of a sample based on initial data associated with said region. For example, the materials that are potentially present in the region of the sample may be determined by assigning a score to the likelihood that the initial sample data indicates the presence of the material, and then selecting a set of one or more potential materials who have the highest score (i.e., the highest three scoring materials), that exceed a threshold value, or a combination thereof. In another example, the potential materials may be determined based on a user input and/or accessible data structure that identifies potential materials that could be present in the type of sample, groupings of materials that are commonly confused with each other, and/or materials of interest, and then select the one or more potential materials based on the information in the data structure.
At 206, a differentiation detector window is identified. The differentiation detector window is a detector window that, when the associated region of the sample is scanned while a detector has the differentiation detector window, the sample data obtained is from a reduced information window that includes a differentiating feature of the one or more potential materials determined in step 204. The differentiation detector window may be a preset windows for differentiating between one or more of the potential materials identified in step 204, or may be determined based on a comparing of the spectral fingerprints associated with each of the potential materials to identify spectral ranges and/or detector windows that capture at least one difference that allows the materials to be differentiated. Then, at 208, the detector settings of the detector are adjusted to obtain new sample data within the differentiation detector window.
At 210, the sample is scanned with a charged particle beam. In some embodiments, scanning the sample comprises scanning the region of interest alone, scanning a large portion of the sample including the region of interest, and/or scanning multiple regions on the sample that have a similar and/or same set of potential materials present.
At 212, it is determined whether an additional region of interest on the sample is to be scanned. If the answer is no, then the process continues to step 214, and the actual materials present in the regions of interest on the sample are determined. Specifically, based on the sample data obtained while each region was scanned while the detector had an associated differentiation detector window, it is determined whether differentiating features are present in the sample data such that one of the potential materials can be either verified as being present or being absent from the associated region of the sample.
If the answer at step 212 is yes, then it is determined at step 216 whether a new differentiation detector window needs to be identified for the additional region of interest on the sample. If the answer at step 216 is yes (i.e., the additional region of the sample has different potential materials or has unknown potential materials), then the process returns to step 204 and one or more new potential materials are determined as potentially being present in the additional region of interest. In this way, an appropriate differentiation detector window can be identified for the additional region of the sample. Alternatively, if it is determined at step 216 that a new differentiation detector window does not need to be determined (i.e., the additional region has the same set of potential materials, the same differentiation detector window, or a different differentiation detector window that has been previously calculated) then the process returns to step 208, where the detector settings are adjusted to the new differentiation detector window (if necessary), and then the process continues to step 210 where the additional region of interest is scanned.
A person having skill in the art would understand that the order of steps shown in
Initial image 302 corresponds to a low resolution image of a sample comprising Cerium and Iron atoms that was generated using HAADF detector data. Such initial sample data is used to determine potential materials/characteristics of materials located at different regions of the sample. For example, the initial information in images 302 can be used to identify the regions in which atoms are located, an estimated size of the atoms in each region, similarity between the shown atoms in different regions, etc. Additionally, image 304 depicts a low resolution EELS spectrum of the sample shown in 302. In various embodiments, the spectrum in 304 may have been obtained by the same system (e.g., charged particle system 104), sequentially, simultaneously, by different systems. The EELS spectrum shows the presence of Oxygen, Iron, and Cerium in the sample. By combining the information in image 302 and spectrum 304, a mapping of the potential materials at various locations of the sample can be created.
Image 306 shows a sample mask that identifies a first set of regions having vertical stripes within which the low resolution EELS data showed the likely presence of Cerium, and a second region having diagonal stripes which indicates a portion of the sample that the EELS data showed was likely to be composed of Iron. The mask identifies the regions of the sample that are to be irradiated while the detector has the associated differentiation detector window to generate detector data to determine the material present in the location. The regions that are not highlighted in the mask are either not to be irradiated, or are to be briefly irradiated while the beam is moving between regions of interest. Specifically, image 306 shows a zoomed in subsection of the sample depicted in image 302.
In 310, the image shows differentiation detector windows that have been identified as allowing both the upper two spectral fingerprints to be easily differentiated from the lower two spectral fingerprints (i.e., the lower two fingerprints have not spectral information associated with their fingerprints), while also allowing the upper two spectral fingerprints to be differentiated from each other (i.e., the first spectral fingerprint has a peak at a lower wavelength than the second spectral fingerprint). By narrowing the detector settings so that it only detects sample data within the differentiation detector window, noise from the remaining portions of the spectrum are not captured, so the sample data acquired is optimally captured to determine the actual material present in the associated location. For example, the leftmost depicted differentiation detector window shows the spectral range within which EELS spectrum data from Ce3+ can be differentiated from Ce4+. Thus, by changing the detector settings to optimally obtain information within this DDW, the detector is able to efficiently capture only the data required to make this differentiation without capturing superfluous information. This allows the determination to be made with reduced beam dose and beam irradiation times.
Finally, image 314 shows a version of image 306 that has been overlaid with an identification, based on the sample data obtained using the beam path shown in image 312 and the differentiation detector windows shown in 310. As can be seen, the central sliver was determined to be composed entirely of Iron atoms of a single oxidation state, while the upper and lower regions identified in mask 306 have been determined to be composed of a Ce of a first oxidation state near the left edge and Ce of a second oxidation state on the right inward side of the sample.
In
In
Image 410 shows a customized scan strategy for the sample that is generated based on the combined mask 406. The scan strategy 410 indicates a beam path that the charged particle is to traverse when it irradiates the surface of the sample. While the scan strategy shown in 410 is generic, a person having skill in the art would understand how a scan strategy would be derived from the mask 406 and differentiation detector windows 410 according to the present invention. Image 410 also shows points in each respective region of interest that the beam is to have a larger dwell time so that detector data for the region of interest can be obtained. Additionally, unlike the scan paths shown in
Examples of inventive subject matter according to the present disclosure are described in the following enumerated paragraphs.
